The Therapeutic Effects of Empathy in Healthcare


Professor Jodi Halpern

Jodi Halpern, School of Public Health, University of California, Berkeley


July 2017 – Increasing attention is being paid to the role of empathy in healthcare, but too little is understood about which specific aspects of empathy are therapeutic and how those aspects work together in clinical encounters.

Traditionally, physicians believed that they could employ a special kind of “detached concern” in which they label patients’ emotions from the outside, looking in, but need not experience actual empathy in which they vividly imagine what patients are going through. They argued that this detachment was essential for them to be objective and not burn out.

Starting about 20 years ago, scholarly work began to emerge that argued against these assumptions and investigated the therapeutic value of emotional empathy (resonating with patients’ feelings and attuning to them non-verbally) as well as cognitive empathy (seeing things from the patients’ perspective as opposed to one’s own).

In From Detached Concern to Empathy, I argued for an integrated affective-cognitive model for clinical empathy in which affects guide what the physician is able to imagine about the patient’s experience (Halpern, 2001). This work was followed by empirical research (reviewed below) showing that the physician’s combined affective-cognitive engagement does increase the effectiveness of medical care.

This essay considers the clinical benefits of distinct aspects of clinical empathy – affective resonance, perspective-taking, compassion, imagining how – and presents a model of how these aspects work in an integrated way to yield the most effective empathic communication. I also address concerns about empathy and burnout and argue for an “empathic curiosity” model of clinical empathy that may help practitioners avoid some of the risks of burnout that have been attributed to “sympathetic distress.” 

The Evolution of Empathy Research

Increasing conceptual precision and improving empirical research on empathy are recent occurrences. A review of hundreds of earlier studies of empathy in healthcare reveals many different uses of the term “empathy” and very little precision in doing so (Pedersen, 2009). In addition, most of the empirical studies, unfortunately, asked doctors to self-report the degree to which they perceive themselves as generally empathic or even as having listened to a patient empathically – not necessarily a valid representation of their actually having engaged empathically.

In addition, even the higher quality research on empathy (psychology research outside of the medical setting) has tended to dichotomize affective and cognitive empathy by providing strictly affective or strictly informational stimuli (Zaki, Bolger, Ochsner, 2008). Yet patients undeniably bring both emotions and thoughts to their encounters with physicians, and for reasons I present below, it is crucial that they be met with empathy that integrates affective and cognitive elements.

Fortunately, over the past couple of decades the quality of empirical research on clinical empathy has improved, with more precise hypotheses and with improved methods that include direct observational studies in which doctor–patient interactions are videotaped over a year of clinical encounters (so that both parties adapt to videotaping in the clinic), as well as MRI and other studies (Decety, Smith, Norman, & Halpern, 2014; Finset, 2011; Suchman, Markakis, Beckman, & Frankel, 2011).

This is an exciting time to return to foundational questions about how empathy is therapeutic because we need to develop more accurate models of how emotional and cognitive aspects of empathy work together to create the dynamic, interpersonal communication that is essential for effective healthcare. That is, we need to get beyond academic questions about what goes on inside the mind of the empathizer in order to take a patient-centered perspective and address how emotional and cognitive features enable patients to feel understood, develop trust, and experience greater agency – all of which are demonstrably therapeutic.

The Individual Aspects of Clinical Empathy and their Therapeutic Contributions

To begin investigating how the affective and cognitive aspects of empathy are related in the complex process of empathic engagement in a health care setting, it is important to delineate the four different aspects of clinical empathy. Their division into separate elements is, however, artificial, for in fact, they are usually (either wholly or partly) integrated. Some of them are interdependent, and all of them can enhance the others – hence “aspects,” implying elements of a larger, unified phenomenon.

The first aspect is affective resonance, which is also related to the more general phenomena of non-verbal attunement. Resonance is the familiar experience of a listener feeling an emotion, often at a reduced or subtle level, that corresponds to that of the speaker. This type of empathy has been shown, empirically, to play an important role in effective healthcare. For example, it can be crucial for enhancing history-taking and thus also diagnosis. Replicated observational studies of patient–physician interactions have shown that before patients talk about aspects of their history that are emotional, they give hints, often through gestures (Suchman et al., 1997). Clinicians who are non-verbally attuned respond to those hints, and then patients communicate (Finset, 2011). As an example, consider a woman oncologist whose young patient seems anxious not only about her own health after breast cancer surgery but also about something else. The oncologist feels a low level of “contagious” worry and communicates this to the patient non-verbally with raised eyebrows, expressing that she is not only sorry for what the patient is going through but is resonating with the patient’s feeling of anxiety. The patient discloses that she is worried about how all this will affect her marriage.

The second aspect of clinical empathy is what psychologists commonly call perspective-taking. This is seen in psychology research as a basic cognitive capacity to perceive a situation from another person’s point of view. There has been little research on perspective-taking in healthcare, but it would seem to play a crucial role in helping patients with decision-making as well as other crucial aspects of healthcare. Research indicates that a lack of perspective-taking is one of the main triggers causing patients to file malpractice claims (Virshup, Oppenberg, & Coleman, 1999). In fields outside of healthcare, perspective-taking has been shown to play a core role in conflict resolution (Goldstein, Vezich, & Shapiro, 2014). Ethics and psychiatry consultants, who listen to countless cases in which patients are unhappy with medical care, understand that helping physicians consider the patient’ perspective is crucial for problem solving.

The claim that perspective-taking can reduce problematic breakdowns in patient–physician communication must acknowledge an inherent challenge, however: it is harder to take on another person’s perspective during a conflict, so getting to the beneficial effects of perspective-taking requires the additional skill of being able to stay interested in a point of view in conflict with your own. A related complication is that it can also be harder for physicians to perspective-take if resonance devolves into sympathetic distress. For example, medical students who become more personally distressed in response to patients’ distress have been shown to have steeper declines in cognitive empathy as their training progresses (Neumann et al., 2011). Even clinicians who stay emotionally engaged can sometimes find themselves feeling sympathetic distress, leading to outcomes that are less therapeutic.

In the example above, what role does perspective-taking play in empathic communication between the oncologist and the young woman with breast cancer?  Because the young woman’s breast cancer is at a very early stage, surgery alone offers a greater than 95% chance of avoiding recurrence, so the oncologist believes the patient need not go through noxious radiotherapy. But the patient disagrees, and the oncologist accepts that for this particular patient even a 5% chance of recurrence is unacceptable. By being able to understand her patient’s point of view, the oncologist will be able to help her make difficult decisions.

The third aspect of clinical empathy is caring, or compassion, feeling a mixture of good will towards and a desire to offer help (as opposed to pity) to the patient. Clinicians’ express a range of benevolent emotions from concern or even worry to mindful compassion. There is increasing evidence that emotionally engaged clinicians – who patients perceive are genuinely worried about them when the situation warrants it– have greater therapeutic efficacy (Roter et al., 1998). This perceived emotional concern engenders trust, which is the most important predictor of adherence to treatment.  Since about half of medical regimens are not followed, causing poor outcomes, improving adherence greatly improves the effectiveness of medical care.  The empathic oncologist in our example conveys both her compassion – as a woman and fellow human she feels with the patient for having to face a serious health problem so young – and she feels and conveys a desire to help her and not abandon her.

The final process in the exercise of clinical empathy is imagining how an experience feels when one is inside it. This is the art of “imagining how…” rather than “knowing that.” My own work draws from the psychodynamic aspects of medical practice and from palliative care to emphasize that empathy involves the act of imagining what is significant from another person’s perspective (Halpern, 2001). Imagining how is guided by curiosity to know what it feels like to be inside the patient’s situation, and makes use of affective resonance to guide what the listener imagines. Imagining how someone else feels is like a daydream or fantasy, insofar as it has not only specific details, but also a connecting theme or mood that conveys that this is a happy, sad, exciting, frightening or other emotional experience.  If a friend tells you that her spouse gave her a vacuum cleaner for Valentine’s Day, you will immediately imagine a very different scenario if you pick up on enthusiastic feelings or if you resonate with hurt feelings.  In the clinical setting, resonance also helps patients give more information, as noted above, making it more likely for the listener to be able to fill in a more vivid picture of what the patient’s concerns are really about.

How does curiosity and trying to imagine the experience of the patient help therapeutically? Insofar as the practitioner can better imagine what is salient for the patient, from that patient’s perspective, she is more likely to get information that is crucial for determining effective and appropriate treatment. Second, it can help with diagnosis. (Why can’t this patient get out of bed all day? Is the apparent lethargy depression or hypothyroidism or a cancer? What comes first, fatigue or demoralization?) Third, both effective medical care and appropriately helping patients with important decisions depend upon it. (Why doesn’t this person with chronic depression take antidepressants? Why is this patient seeking surgery for back pain despite poor odds that it will help? Why does this person seek aid in dying?) Vividly imagining a patient’s emotional experience goes beyond cognitive perspective-taking insofar as it is guided by resonance and thus feels like something affective is happening to the listener—she is actually daydreaming the patient’s world and not just knowing facts about it.

How do practitioners take in enough of the details of the patient’s world to be able to construct at least a momentary experiential grasp, or daydream-like experience? First, they have to be good listeners. Second, they have to develop the trust that allows patients to tell them more. We have already seen that resonance helps develop trust. So resonance enables patients to give more details, which then enables the doctor to picture things more fully, which she then can convey back to the patient, which in turn develops more trust. This is a virtuous cycle. The details gathered may not be particularly accurate at first, but as long as the practitioner lets the patient see that she is trying to understand things in detail rather than brush over them, and if she can invite the patient to point out mistakes and make corrections, the best empathic communication can happen.

To return to the example of the oncologist: she knows that her highly educated patient understands the medical facts, including how noxious the chemotherapy will be and that it introduces long term health risks. She is curious about how this patient’s palpable fear may be contributing to her determination to pursue the most aggressive treatment. Based on their communication, she is able to imagine being inside the young woman’s experience and comes to understand that the patient feels as if she’s been cursed. She believes that having been unlucky enough to get cancer so young she will be unlucky again and end up in the 5% recurrence group. The oncologist helps the patient become conscious that this feeling of being singled out by fate is a common manifestation of fear. In grasping this, the patient begins to cry over the whole situation and eventually realizes that she cannot fully control the uncertainties in her future. She reconsiders her initial treatment decision.

Empathic Curiosity Enhances Therapeutic Benefits

The empathic curiosity model that I propose for the practice of clinical empathy involves all four of the aspects that I have outlined. It puts resonance to work in the service of learning more in order to better imagine how the patient is feeling and to communicatively connect in real time while co-imagining, in order to convey caring and appropriate worry.

Why is this model distinct from and of much greater therapeutic value than detached perspective-taking in clinical care?

First, being approached with detached curiosity makes people who are hearing bad news or are otherwise suffering feel like they are disappearing (Brison, 1996; Girgis & Sanson-Fisher, 1998; Ptacek, Fries, Eberhardt, & Ptacek, 1999). In contrast, empathic curiosity builds trust and is empowering (Halpern, 2001; Roter et al., 1998). Second, the model, as a whole, has a positive “cyclical” effect: affective engagement builds trust / resonance guides what the listener imagines / attuning of mood enables more relevant imagining how / imagining how improves communication in a virtuous cycle. In our case example, the patient senses the oncologist’s affective resonance and that builds trust; and for her part, the oncologist’s resonance with the patient’s anxiety leads her to imagine an anxious world view rather than a sad or a catastrophic one. All of this helps them better align their points of view and thus communicate more effectively.

Empathic curiosity and compassion supplement each other in empathic therapeutic encounters. On the one hand, curiosity without compassion or a pro-social attitude towards the patient could be used for dangerous or destructive purposes. On the other hand, generic compassion for all fellow humans lacks the curiosity about what, specifically, this patient is concerned about.  These two situations represent degraded forms of empathic curiosity and compassion. In full empathic communication, on the other hand, the doctor’s empathic curiosity about the individual patient is guided by a sense of shared humanity and shared possibilities for vulnerability and finitude.

Given that accepting one’s vulnerability and finitude is not easy, the newer emphasis on training doctors in “mindful compassion” is a good step in the direction of facilitating empathy. Some hospitals, however, now translate “compassion” into “be kind,” which devalues it and risks a return of paternalistic projections of “kindness,” as exemplified by the phrase too often uttered by uncurious doctors: “I know how you feel.” Empathic curiosity, in contrast, respects the individuality of each patient and says instead: “Tell me what I’m missing.” Even a woman doctor who has had her own breast cancer and who thinks she knows how to be kind to a woman patient with breast cancer is at risk of missing important particulars of the patient’s world.

One might wonder: What about avoiding burnout? In my view, the best path is to cultivate both empathic curiosity and mindful compassion, as the two work synergistically, as described above. We have enough research to take seriously the idea that compassion practices can help avoid burnout. We need more precise research on empathic curiosity, but we do have suggestive correlational findings that curiosity about another person’s feelings helps caregivers reduce self-related anxiety (Decety et al., 2014). And all of us are familiar with the fact that becoming engaged with and curious about another person’s experience is the surest way to decenter from our own anxiety – consider how people relax by reading or watching television or film narratives of others’ lives. My decades of clinical observations and educating medical and nursing students have shown that those who were able to sustain interest in their practices were the ones who stayed curious about what their own emotional responses to difficult clinical encounters might be telling them about their patients. Of course this requires not only curiosity but also self-awareness, which is where the mindfulness component can be very useful. That developing curiosity and one’s empathic imagination reduces personal distress is reflected in a study that found that oncologists and hospice physicians with “exquisite empathy” showed the lower levels of burnout and distress than other physicians (Kearney, Weininger, Vachon, Harrison, & Mount, 2009).

For physicians to be mindful when they themselves are anxious helps the situation not devolve into over-identification, projection, or feeling helpless. On the other hand, when a doctor does feel helpless or merged, it is much better if she can recognize it and get help.  Physicians who are unaware of their upset feelings are at risk of creating problems through everything from precipitous discharges to over or under treatment (Halpern, 2001).

In summary, the empathic curiosity model incorporates aspects of compassion in that it integrates a pro-social concern or feeling for the well-being of another with ongoing imagination work that attempts to grasp what the other person’s individual, subjective experience feels like from the inside out.

In Conclusion

Given this model, and the demonstrable effectiveness of its components, the question that follows is how to create an organizational culture that motivates empathic curiosity and emotional engagement so that patients can receive empathic care?  Recent work by Jamil Zaki and others suggests how social expectations and environmental cues can readily motivate empathy or the lack of it (see Zaki, 2014). One important and underexplored area of social motivation is the expectations of patients who have been educated and empowered collectively to expect empathy.  An interdisciplinary group of us recently argued for the importance of expanding our research agenda beyond the longstanding focus on the intra-psychic aspects of empathy to include the dynamic, interpersonal aspects of successful empathic communication (Main, Walle, Kho, & Halpern, 2017). Achieving this would suggest the need to develop intervention studies to see if, in addition to educating physicians, empowering patients to expect and skillfully trigger empathy improves patient–physician communication. In this regard, we also need to work to eliminate systemic barriers to empathy, which include not only racism, sexism, gender and sexual orientation biases, and able-ism but also the specific devaluing of people with obesity, addictions or just “unhealthy” habits.

Over a decade ago I made a philosophical argument that when affective and cognitive empathy are integrated, the results are especially therapeutic (Halpern, 2001), and the empirical research that has been done since supports this. I hope this essay can help to bring more precision to the field of clinical empathy by uniting empirical findings and conceptual arguments in order to delineate testable hypotheses regarding which specific aspects of empathy are therapeutic. I also hope it helps to call into question the current shift in healthcare toward “feeling for” patients (compassion) and away from “feeling into” them (empathy). The rise of mindfulness and compassion need not be accompanied by the devaluing of empathic curiosity – in fact I hope I’ve shown that best practices integrate these approaches. Further, I continue to build on my hypothesis that the most powerful therapeutic/mutative factor in all the aspects of clinical empathy is rooted not in another person having compassion for you but in another person knowing how it feels to be in your world from the inside out. I hope to see both more precise research testing this hypothesis and the development of interventions to help physicians become better at empathic communication and imagining how it feels to be in their patient’s shoes.



Brison, S. J. (1996). Outliving oneself: Trauma, memory and personal identity. In D. Meyers (Ed.), Feminists Rethink the Self (pp. 13-45). Boulder, CO: Westview Press.

Decety, J., Smith, K., Norman, G., & Halpern, J. (2014). Clinical empathy: What can we learn from social and affective neuroscience? World Psychiatry, 13, 233-237.

Finset, A. (2011). Research on person-centered clinical care. Journal of Evaluation in Clinical Practice, 17, 384-386.

Girgis A., & Sanson-Fisher, R. W. (1998). Breaking bad news 1: Current advice for clinicians. Behavioral Medicine, 24, 53-59.

Goldstein, N. J., Vezich, I. S., & Shapiro, J. R. (2014). Perspective-taking: When others walk in our shoes. Journal of Personality and Social Psychology, 106, 941-960.

Halpern J. (2001). From Detached Concern to Empathy: Humanizing Medical Practice. Oxford, UK: Oxford University Press.

Kearney, M. K., Weininger, R. B., Vachon, M.S., Harrison, R. L., Mount, B. M. (2009). Self-care of physicians caring for patients at the end of life: ‘Being connected . . . A key to my survival. Journal of the American Medical Association, 301, 1155-1164.

Main, A., Walle, E. A., Kho, C., Halpern, J (2017). The interpersonal functions of empathy: A relational perspective. Emotion Review.

Neumann, M., Edelhäuser, F., Tauschel, D., Fisher, M. R., Wirtz, M., Woopen, C., Haramati, A., & Scheffer, C. (2011). Empathy decline and its reasons: A systematic review of studies with medical students and residents. Academic Medicine, 86, 996-1009.

Pedersen, R. (2009). Empirical research on empathy in medicine – A critical review. Patient Education and Counseling, 76, 307-322.

Ptacek, J. T., Fries, E. A., Eberhardt, T. L., Ptacek, J. J. (1999). Breaking bad news to patients: Physicians’ perceptions of the process. Support Care Cancer, 7, 113-120

Roter, D. L., Hall, J. A., Merisca, R., Nordstrom, B., Cretin, D., & Svarstad, B. (1998). Effectiveness of interventions to improve patient compliance: A meta-analysis. Medical Care, 36, 1138-1161.

Suchman, A. L., Markakis, K., Beckman, H. B., Frankel, R. (1997). A model of empathic communication in the medical interview. Journal of the American Medical Association, 277, 678-682.

Virshup, B. B., Oppenberg, A. A., Coleman, M. M. (1999). Strategic risk management: Reducing malpractice claims through more effective patient-doctor communication. American Journal of Medical Quality, 14, 153-159.

Zaki, J. (2014). Empathy: A motivated account. Psychological Bulletin, 140, 1608–1647.

Zaki, J., Bolger, N., Ochsner, K. (2008). It takes two: The interpersonal nature of empathic accuracy. Psychological Science, 19, 399-404.

Print Friendly, PDF & Email

Traffic Count

  • 312147Total reads:
  • 782Reads today:
  • 251402Total visitors:
  • 440Visitors today: